PretrainedAligner¶
- class aligner.aligner.PretrainedAligner(corpus, dictionary, acoustic_model, output_directory, beam=100, temp_directory=None, num_jobs=3, speaker_independent=False, call_back=None, debug=False, skip_input=False)[source]¶
Class for aligning a dataset using a pretrained acoustic model
- Parameters:
- corpus
Corpus
Corpus object for the dataset
- dictionary
Dictionary
Dictionary object for the pronunciation dictionary
- acoustic_model
AcousticModel
Archive containing the acoustic model and pronunciation dictionary
- output_directorystr
Path to directory to save TextGrids
- temp_directorystr, optional
Specifies the temporary directory root to save files need for Kaldi. If not specified, it will be set to
~/Documents/MFA
- num_jobsint, optional
Number of processes to use, defaults to 3
- call_backcallable, optional
Specifies a call back function for alignment
- corpus
Attributes
meta
mono_ali_directory
mono_directory
mono_final_model_path
tri_ali_directory
tri_directory
tri_final_model_path
tri_fmllr_ali_directory
tri_fmllr_directory
tri_fmllr_final_model_path
Methods
do_align
()Perform alignment while calculating speaker transforms (fMLLR estimation)
export_textgrids
()Export a TextGrid file for every sound file in the dataset
get_num_gauss_mono
()Get the number of gaussians for a monophone model
parse_log_directory
(directory, iteration)Parse error files and relate relevant information about unaligned files
setup
()test_utterance_transcriptions
()train_tri_fmllr
()Perform speaker-adapted triphone training